Combining Deep Reinforcement Learning With Graph Neural Networks for Optimal VNF Placement

Network Function Virtualization (NFV) technology utilizes software to implement network function as virtual instances, which reduces the cost on various middlebox hardware. A Virtual Network Function (VNF) instance requires multiple resource types in the network (e.g., CPU, memory). Therefore, an efficient VNF placement policy should consider both the resource utilization problem and the Quality of Service (QoS) of flows, which is proved NP-hard. Recent studies employ Deep Reinforcement Learning (DRL) to solve the VNF placement problem, but existing DRL-based solutions cannot generalize well to different topologies. In this letter, we propose to combine the advantage of DRL and Graph Neural Network (GNN) to design our VNF placement scheme DeepOpt. Simulation results show that DeepOpt outperforms the state-of-the-art VNF placement schemes and shows a much better generalization ability in different network topologies.

Automated summary

NFV technology uses software to implement virtual instances of network functions, reducing costs on middlebox hardware. VNF instances require multiple resource types and considering both resource utilization and QoS is NP-hard. DeepOpt, combining DRL and GNN, outperforms existing VNF placement schemes.